|Publication number||US7403958 B2|
|Application number||US 11/039,472|
|Publication date||Jul 22, 2008|
|Filing date||Jan 19, 2005|
|Priority date||Jan 19, 2005|
|Also published as||US20060161606|
|Publication number||039472, 11039472, US 7403958 B2, US 7403958B2, US-B2-7403958, US7403958 B2, US7403958B2|
|Inventors||Dilip Kumar BISWAL, Isaac Kam-Chak Cheng, Cecilia O. Chu, Donald Lee Clare, Louis S. W. Mau|
|Original Assignee||International Business Machines Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (30), Referenced by (5), Classifications (13), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application is related to co-pending U.S. patent application Ser. No. 11/039,528, entitled “Redundant Version Information in History Table That Enables Efficient Snapshot Querying”, filed concurrently herewith.
The present invention relates to data replication and synchronization, and more particularly to the simultaneous replication and synchronization of data.
One conventional approach for facilitating replication and synchronization is to use a single snapshot of the source 101, shared by both the synchronization and the replication. However, while the data is being written during the replication, any attempts to simultaneously read the data during synchronization are blocked to maintain the integrity of the data retrieved. “Simultaneous” as used in this specification, refers to instructions within two different threads attempting to be executed concurrently. In another conventional approach, a plurality of snapshots is shared while providing isolation levels in the mirror 102. Although this approach allows the simultaneous write (during replication) and read (during synchronization) of the snapshots, simultaneous writes during replication and synchronization are blocked. Thus, the conventional approaches are limited in their efficiency of the replication and synchronization processes.
Accordingly, there exists a need for an improved method and system for synchronization-replication concurrency. The method and system should allow simultaneous reads, writes, and read-writes during replication and synchronization, while maintaining data integrity. The present invention addresses such a need.
An improved method and system for synchronization-replication concurrency maintain a history table for a data table in the mirror 102, where the history table can include redundant version values that indicate a period in which data values corresponding to the version values are valid. Replication can thus be performed between a source and a mirror for a current version of the data table, while synchronization is simultaneously performed between the mirror and a client for an earlier version of the data table using the history table. Simultaneous reads, writes, and read-writes are possible while maintaining data reliability. The replication and synchronization processes can be performed at the read-uncommitted isolation level to increase efficiency. The method and system thus increase overall data throughput, reduce client response time, and dramatically enhance end-user satisfaction.
The present invention provides an improved method and system for synchronization-replication concurrency. The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a patent application and its requirements. Various modifications to the preferred embodiment will be readily apparent to those skilled in the art and the generic principles herein may be applied to other embodiments. Thus, the present invention is not intended to be limited to the embodiment shown but is to be accorded the widest scope consistent with the principles and features described herein.
To more particularly describe the features of the present invention, please refer to
Although the embodiment above is described in the context of a history table that uses redundant version values, one of ordinary skill in the art will understand that other types of history tables may be used without departing from the spirit and scope of the present invention.
At version 7, the history table indicates that the data value ‘10’ for key ‘1’ is valid from version 1 onward (represented by infinity), the data value ‘6’ for key ‘2’ is valid from version 1 to 5, and the data value ‘20’ for key ‘2’ is valid from version 6 onward. At version 8, a row update is performed for key ‘2’ to update the data value to ‘888’. Correspondingly, a row for key ‘2’ and data value ‘888’ is inserted into the history table, with its ‘from’ value set to ‘8’ and the ‘to’ value set to infinity. The old row for key ‘2’ with data value ‘20’ is updated so that the ‘to’ value is set to ‘7’. This indicates that for key ‘2’, the data value ‘20’ is valid from version 6 to 7, while the data value ‘888’ is valid from version 8 onward. Also at version 8, a row insert is performed on the data table to insert key ‘3’ with data value ‘30’. Correspondingly, a row for key ‘3’ and data value ‘30’ is inserted into the history table, with its ‘from’ value set to ‘8’ and the ‘to’ value set to infinity. This indicates that for key ‘3’, the data value ‘30’ is valid from version 8 onward.
Assume that replication is performed at version 8, and synchronization is performed at version 7. As illustrated in
With the method and system in accordance with the present invention, no new row can be introduced into and no row can be deleted from the previous version of the data table used by synchronization during replication of the current version. This is because any new row introduced by the replication would have a ‘from’ version value in the history table greater than the version for synchronization. These rows in the history table would not be retrieved by the snapshot query. Also, the ‘from’ version value for old rows in the history table is never changed. Thus, a snapshot of a previous version will be reliably retrieved by the snapshot query. This is true even at the read-uncommitted isolation level. The simultaneous replication and synchronization can thus be performed at this isolation level, increasing the efficiency of the processes, while also ensuring data reliability.
An improved method and system for synchronization-replication concurrency have been disclosed. The method and system maintain a history table for a data table, where the history table includes redundant version values that indicate a period in which data values corresponding to the version values are valid. Replication can thus be performed between a source and a mirror for a current version of the data table, while synchronization is simultaneously performed between the mirror and a client for an earlier version of the data table using the history table. Simultaneous reads, writes, and read-writes are possible while maintaining data reliability. The replication and synchronization processes can be performed at the read-uncommitted isolation level to increase efficiency. The method and system thus increase overall data throughput, reduce client response time, and dramatically enhance end-user satisfaction.
Although the present invention has been described in accordance with the embodiments shown, one of ordinary skill in the art will readily recognize that there could be variations to the embodiments and those variations would be within the spirit and scope of the present invention. Accordingly, many modifications may be made by one of ordinary skill in the art without departing from the spirit and scope of the appended claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5623661||Nov 29, 1995||Apr 22, 1997||International Business Machines Corp.||System for and method of providing delta-versioning of the contents of PCTE file objects|
|US5675802||Mar 31, 1995||Oct 7, 1997||Pure Atria Corporation||Version control system for geographically distributed software development|
|US5870758||Mar 11, 1996||Feb 9, 1999||Oracle Corporation||Method and apparatus for providing isolation levels in a database system|
|US5870759 *||Oct 9, 1996||Feb 9, 1999||Oracle Corporation||System for synchronizing data between computers using a before-image of data|
|US5950210||Jan 29, 1997||Sep 7, 1999||Nelson; Paul M.||Database row version differentiation process|
|US5956713||Apr 23, 1997||Sep 21, 1999||Oracle Corporation||Techniques for reducing the number of snapshots of a database|
|US5956731||Apr 23, 1997||Sep 21, 1999||Oracle Corporation||Sharing snapshots for consistent reads|
|US5963959||May 30, 1997||Oct 5, 1999||Oracle Corporation||Fast refresh of snapshots|
|US5999931||Oct 17, 1997||Dec 7, 1999||Lucent Technologies Inc.||Concurrency control protocols for management of replicated data items in a distributed database system|
|US6092086||Mar 31, 1998||Jul 18, 2000||Bmc Software||System and method for handling backout processing during capture of changed data in an enterprise computer system|
|US6289335||Jun 23, 1997||Sep 11, 2001||Oracle Corporation||Fast refresh of snapshots containing subqueries|
|US6295541 *||Aug 18, 1998||Sep 25, 2001||Starfish Software, Inc.||System and methods for synchronizing two or more datasets|
|US6377960 *||Jul 26, 1999||Apr 23, 2002||Microsoft Corporation||Transactional configuration store and runtime versus administration isolation with version snapshots and aging|
|US6415299||Apr 12, 2000||Jul 2, 2002||Unisys Corp.||Method for merging versions of a model in an object oriented repository|
|US6460052||Aug 20, 1999||Oct 1, 2002||Oracle Corporation||Method and system for performing fine grain versioning|
|US6516327 *||Sep 24, 1999||Feb 4, 2003||International Business Machines Corporation||System and method for synchronizing data in multiple databases|
|US6631374||Sep 29, 2000||Oct 7, 2003||Oracle Corp.||System and method for providing fine-grained temporal database access|
|US6665815||Jun 22, 2000||Dec 16, 2003||Hewlett-Packard Development Company, L.P.||Physical incremental backup using snapshots|
|US6799189||Nov 15, 2001||Sep 28, 2004||Bmc Software, Inc.||System and method for creating a series of online snapshots for recovery purposes|
|US20020029218||Oct 22, 2001||Mar 7, 2002||Bentley Keith A.||System, method and computer program product for collaborative engineering using component and file oriented tools|
|US20020174416||Dec 6, 2001||Nov 21, 2002||International Business Machines Corporation||Storing and restoring snapshots of a computer process|
|US20020178146||May 24, 2001||Nov 28, 2002||International Business Machines Corporation||System and method for selective object history retention|
|US20030131025 *||Dec 20, 2002||Jul 10, 2003||Zondervan Quinton Yves||System and method for synchronizing data in multiple databases|
|US20030149702||Feb 5, 2002||Aug 7, 2003||Compaq Information Technologies Group, L.P.||Operational data store|
|US20040010487||Sep 30, 2002||Jan 15, 2004||Anand Prahlad||System and method for generating and managing quick recovery volumes|
|US20040019614 *||Jul 24, 2002||Jan 29, 2004||International Business Machines Corporation||Mid-tier-based conflict resolution method and system usable for message synchronization and replication|
|US20050086263||Sep 3, 2004||Apr 21, 2005||Oracle International Corporation||Self-managing performance statistics repository for databases|
|US20050131964||Dec 12, 2003||Jun 16, 2005||Oracle International Corporation||Object versioning|
|US20050144198||Feb 11, 2005||Jun 30, 2005||Microsoft Corporation||Versions and workspaces in an object repository|
|US20060101092||Dec 23, 2004||May 11, 2006||Hitachi, Ltd.||Computer system and method for managing file versions|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8290966 *||Nov 29, 2007||Oct 16, 2012||Sap Aktiengesellschaft||System and method for implementing a non-destructive tree filter|
|US8930499||Jul 9, 2010||Jan 6, 2015||International Business Machines Corporation||Data replication between software versions|
|US9124609 *||Mar 13, 2007||Sep 1, 2015||International Business Machines Corporation||Ensuring consistency over time of data gathered by distinct software applications|
|US20090282097 *||Mar 13, 2007||Nov 12, 2009||International Business Machines Corporation||Method and System for Ensuring Consistency Over Time of Data Gathered By Distinct Software Applications|
|US20110289054 *||Nov 24, 2011||Southern Company Services||Content Management Migration Manager System and Method|
|U.S. Classification||1/1, 707/E17.007, 714/E11.128, 707/999.204, 707/999.202|
|Cooperative Classification||Y10S707/99953, Y10S707/99955, H04L67/1095, G06F17/30356, G06F17/30575|
|European Classification||H04L29/08N9R, G06F17/30C|
|May 23, 2005||AS||Assignment|
|Mar 5, 2012||REMI||Maintenance fee reminder mailed|
|Jun 29, 2012||FPAY||Fee payment|
Year of fee payment: 4
|Jun 29, 2012||SULP||Surcharge for late payment|